The separation and collection of particulates/aerosols from an airstream (or other fluid streams) is of concern in several contexts. In some cases, the goal may be to simply remove the particulates/aerosols from the fluid stream, thereby cleaning or purifying the fluid. Often it is desired to remove all particulates, regardless of composition, if the particulates are above a certain size. For example, automobile painting and the fabrication of silicon chips in clean rooms represent two situations in which all particulates large enough to result in an inferior product are desirably removed from the processing environment.
In other cases, particulates are collected for analysis to determine the type and concentration of such particulates/aerosols entrained in the fluid. For example, this technology may be employed in the detection of airborne biological or chemical warfare agents, the detection of biological contamination in confined spaces, such as aircraft or hospitals, or the detection of industrial pollutants (either in ambient fluid or in the effluent of smokestacks).
Much effort has been expended in the past in the detection and classification of particulates or aerosols in fluid streams. Impactors have been used for collecting aerosol particulates for many decades. In the earliest embodiments, a stream of fluid containing the particulates was accelerated toward an impactor plate. Due to their inertia, the particulates striking the impactor plate were collected on its surface, while the fluid was deflected to the side. With these types of impactors, only larger particulates could be collected, since particulates below a certain “cut size” were carried away by the fluid stream.
However, a significant disadvantage of such an impactor is the deposition of particulates on surfaces of the impactor other than the intended collection surfaces. This phenomenon reduces the accuracy of measurement of total particulate mass concentration and of the size-fractionation of particulates, since such losses cannot be accurately estimated for aerosols or particulates of varying size, shape, or chemistry. Additionally, particulates may either become re-entrained in the fluid stream, or may bounce off the impactor's collection surface upon impact. To remedy this problem, “virtual” impactors have been developed that separate particulates from a fluid stream with techniques other than direct impaction. Virtual impactors may operate on a number of different principles, but all avoid actual “impact” as a means to separate particulates from a fluid in which the particulates are entrained and rely on differences in particulate mass to induce inertial separation. Specifically, a particulate-laden fluid stream is directed toward a surface presenting an obstruction to the forward movement of the fluid stream. The surface includes a void at the point where the particulates would normally impact the surface. When a major portion of the fluid stream changes direction to avoid the obstruction presented by the surface, fine particulates remain entrained in the deflected major portion of the fluid stream. Heavier or denser particulates, on the other hand, fail to change direction and are collected in a region of relatively stagnant fluid (a “dead zone”) that is created near the surface. The heavier particulates entrained in a minor portion of the fluid stream enter the void defined by the surface, where they can be captured or analyzed.
Some examples of virtual impactors can be found in U.S. Pat. Nos. 3,901,798; 4,670,135; 4,767,524; 5,425,802; and 5,533,406. Because typical virtual impactors do not actually collect particulates themselves, but merely redirect them into two different fluid streams according to their mass, they are essentially free of the problems of particulate bounce and particulate re-entrainment associated with actual impactor devices. Still, particulate “wall loss,” i.e., unintended deposition of particulates on various surfaces of virtual impactor structures, especially at curved or bent portions, remains a challenge with some designs of virtual impactors, because typically, many stages or layers of virtual impactors are required to complete particulate separation.
An additional aspect of the collection of fluid-entrained particulates, especially with respect to particulates that will be analyzed to determine a type and concentration of particulates, relates to when the collected particulates are to be analyzed. A common practice is to sample a fluid for a period of time, and then analyze the collected sample immediately, or at least as soon as practical. Depending on the nature of the particulates for which the fluid is being sampled, immediate analysis may be required. For example, if chemical or biological agents that pose an immediate health threat are suspected, real time analysis is preferred to enable protective measures to be taken immediately, before irreversible harm can occur. However, there are also many applications, such as routine monitoring of smokestacks and wastewater discharge, in which only a portion of the collected sample might need to be analyzed shortly after collection, while other portions are best archived for later analysis.
Archival samples can be prepared by taking a collected sample and manually splitting that sample into various fractions, including a first fraction to be analyzed relatively soon, and one or more additional portions to be archived for possible later analysis. While archival samples prepared by such a method are useful, the manual nature of such a method is time consuming and costly. Furthermore, during each step in which a sample is handled or manipulated (collection, separation, storage, and analysis), there is a significant chance that the sample will be inadvertently contaminated. It would thus be desirable to provide a method and apparatus that more readily enables archival samples to be prepared, with a minimal risk of contamination.
It should also be noted that the manner in which samples are collected affects the usefulness of the samples for archival purposes. Archival samples are often employed to determine more information about an event occurring at a specific time. For example, archival data collected from a smokestack might be used to determine at what time higher emissions occurred. That time could then be applied to analyze the process and equipment utilizing the smokestack to isolate the factors causing the excess emissions, so that the problem can be corrected. If the archival sample is merely a single sample collected over a 24-hour period, rather than 24 samples collected each hour for 24 hours, then little information can be obtained about when the excess emissions actually occurred, making it more difficult to determine the cause of the excess emissions. It would be therefore be desirable to provide a method and apparatus capable of providing archival samples for successive relatively short sampling periods, and which include time indexing, enabling a specific archival sample to be correlated with a specific time at which the sample was taken.
Accordingly, a need exists to develop a method and apparatus capable of providing time-indexed archival samples with minimal operator effort, and minimal chance of contamination. Such archival samples desirably should include a high concentration of particulates, so that the archival samples are compact and require minimal storage space. Preferably, a virtual impactor that efficiently separates particulates from a fluid stream might be employed to collect the particulates.
Yet another aspect of the collection of fluid-entrained particulates, especially with respect to particulates collected with an impact collector, relates to how the collected particulates are to be analyzed. Most analytical techniques require a liquid sample. Regardless of how effective impact collectors are at removing particulates from a fluid stream (such as air), the collected particulates generally cannot readily be analyzed while remaining deposited on the impact collection surface. It would be desirable to provide a method and apparatus for removing collected particulates from an impact collection surface, and to transfer such particulates to a container that can be utilized to prepare a liquid sample. It would further be desirable to provide an integrated system capable of collecting particulates from a fluid stream using an impact collector, and then transferring the collected particulates from the impact collector to a container suitable for preparing a liquid sample.
It should be noted that when a liquid sample is prepared using collected particulates, the amount of liquid used to prepare the liquid sample plays a significant factor in determining the concentration of the liquid sample. Higher concentration samples generally require less challenging analytical techniques to analyze and are thus preferred. Therefore, it would be desirable for the method and apparatus employed to transfer collected particulates from an impact collection surface to a suitable container utilizing little or no liquid to unduly dilute the sample.